Research On Electrical Identification Based On Ultrasonic Signal And Its Application In Indoor Positioning

With the development of the aging society,ambient assisted living(AAL)technology has been more widely used.In this scenario,various indoor positioning methods can provide position data basis for it to improve the application of AAL technology.However,due to the huge differences in different indoor environments,mainstream positioning methods based on specific deployment infrastructure have problems such as high degree of intervention,difficulty in transformation,and poor universality in home health care scenarios.Therefore,this paper proposes for the first time the application of electrical ultrasound as a new signal of opportunity for indoor positioning,avoiding infrastructure construction,and converting the ubiquitous unreliable positioning reference information in the environment into highly reliable binary sensing.This enhances the universality of positioning solutions,which is of great significance to promoting home health care services and improving the quality of life of users.This paper first introduces various positioning methods based on sound and other indoor positioning solutions,and briefly introduces their pros and cons.Finally,from the perspective of no need for infrastructure deployment and privacy security,it is proposed to use electrical ultrasound as a signal of opportunity for positioning.Due to its long-term stability and non-intrusive characteristics,it can be used as an anchor point and closed-loop detection point,and has great advantages in the field of indoor positioning.1.In terms of model identification,this paper adopts different noise reduction methods such as spectral subtraction and high-frequency interception for the steadystate noise and low-frequency human voice that may be contained in electrical appliances,so as to improve the model’s ability to recognize electrical appliances.This article also compared the performance of different models in the actual data test situation,and finally chose random forest,which achieved an accuracy rate of 82.47%in the actual test.2.In order to improve the credibility of electrical ultrasound as a signal of opportunity in the field of positioning,this paper proposes a post-processing algorithm.By setting multiple parameters such as the minimum initial length and the minimum effective length,while removing transient noise,it is also possible to properly correct the model recognition results and exclude a small amount of outliers,thereby reducing the misjudgment rate.Experimental results show that the algorithm can still have a recall rate of 79.66%on the premise of ensuring a recognition accuracy of 100%.3.In order to solve the problem that the anchor points of electrical appliances are sparsely distributed in the actual environment,this paper combines the ultrasound of electrical appliances with the algorithm of pedestrian dead reckoning,and conducts experiments in an office environment.In addition,combined with the actual positioning scene,the appliance may move,so this paper proposes to use the appliance as a fixed anchor point or a closed-loop constraint in different situations.The experimental results show that the positioning accuracy of the single pedestrian dead reckoning method can be improved by 95.43%by combining the electrical ultrasonic signal recognition.